A weather station anywhere could reliably transmit its data so that farm insurance companies can price their products. A point of sale terminal anywhere could make e-payments. A neighbour could request and pay for government services. A farmer could be advised to start their harvest early because of impending weather change. A village could get critical advice and response information in case of emergency, be it before a weather-related event, during a disease outbreak or after an earthquake.
The capabilities described above are the challenge. So how do we make it happen? It is important to lead technological innovation with the challenge and not with a particular solution type. Solution types come in many shapes and forms. We could, for example, use a mesh of wifi networks in higher density settings, microwave radio links, GSM networks for actual SMS, high altitude drones for lower density areas, or indeed satellites for most sparse areas. Probably we would use a combination of all these things at different times.
The issue that makes satellites particularly interesting is that wifi, GSM, and drones need expensive investments upfront to fund local infrastructure (towers, antennae, hardware and devices distributed in the area in which you need communication). Satellites share the cost over a much wider area and with a mostly non-local infrastructure. A single satellite can cover and service several countries at once. Moreover, they are extremely reliable and can be trusted to deliver critical data. This, in turn, opens the door to sharing the operating costs with businesses that depend on that communication. For example, an agricultural insurance company or commodity trader would be willing to pay for a satellite service, a premium channel, but one that ensures the data is delivered.
This approach is what “impact science” means for development. That is, to use the scientific method to make sure we understand the challenges and the solutions. The scientific process needs an observation, a hypothesis, and a method to test our assumption and learn the impact so we can adjust if needed or scale the solution if it worked. Our hypothesis is that aiming for broadband access risks leaving many behind in a digital divide. Instead, an approach aiming for a resilient basic connection would create a minimum bottom line from which other use cases can develop. This connectivity assurance bridges the digital divide even for the most isolated places (half a billion people live outside mobile network coverage) and would ensure that those digital dividends can reliably reach more people in more places, allowing these communities to benefit from other technological innovations that depend on connectivity.
Looking at the latest market reports (e.g. from SIA), however, the biggest incentive for satellite communications providers seems to come from elsewhere. There is a huge push, for example, to offer fast wifi in transoceanic aircrafts and, from some deep pockets of eager tech investors with very opinionated use cases, initiatives like Internet.org by Facebook, where some websites and uses are free and some are not. We also have tremendous endeavours like the SpaceX proposal or OneWeb, whose upfront and ongoing operating costs need an extremely fast growth in demand to make the math work.
I would, however, argue that tapping into digital dividends not only helps the most people but also greatly outweighs the size of the markets these main drivers seek. The size of the on-board wifi market is expected to reach $33 billion by 2020. The market for money e-transfers in developing countries is already $138 billion today and 10% of the world still lacks access to data communication (SMS or better).
Basic doesn’t mean simple
A basic connection doesn’t mean that only simple use cases are possible. In fact, this approach connects with many of the latest revolutions of the Fourth Industrial Revolution. We can talk, for example, about autonomous driving (starting with transportation on roads and also by sea) and smart things (Internet of Things devices). In both of these cases, one of the techniques that will be needed is “one-to-many” broadcast services. This means that you send the same data to many users, say a software update, changes in the road network or prices of road tolls. Since the unit only needs to download there is no need to make the receiver complex and powerful to emit a signal back to space. The receiver is then simpler and cheaper.
The interesting bit for development is that many use cases can take advantage of this "broadcasted signal marketplace". In fact, this is the case for many of those examples described above, which involved informing places about emergencies or health pandemics, or the best timing for harvests or market prices - all are examples where you only need to receive information.